Application: An AI-based DSS for Managing Multiple Unmanned Aerial Vehicles

AI algorithms and DSS can be applied to aid in operator decisions for path planning & task allocation of a swarm of drones [1], with the aim of:

• Reduce the operator workload.
• Manual tasks must be automated or simplified.
• Need of a Multi-UAV approach.
• Need of automated Replanning in real-time.
• Need of a Decision Support System (DSS) to facilitate plan selection for operators.
• Improvement of guidance systems with high-level autonomy mission planning, replanning and decision making capabilities.

The Mission Planning Problem

This problem falls under that of the classical Mission Planning Problem (MPP), for optimising task and resource allocation, geographic coordination and timing.

• Nowadays, operators use manually insertion of tasks and waypoints or classic planners.
• Changes can affect the preload plan during its execution, and operator manually replan the mission.

One possible approach to deal with this problem can be based on:

• Artificial Intelligence: Multi-Objective Optimization Problem (MOP), to look for the best optimal solutions (multicritera decision making)
  • Optimise mission cost, risk, fuel consumption, flight time
• Decision Support Systems: to allow improve the perfomance of Ground Control operators
  • Ranking system, based on the operator preferences.
  • Filtering system, based on the similarity of solutions.

Techniques & methodology (4 steps):

1. Constraint Satisfaction Problem (CSP) modelling of the MPP.
2. Resolution of the mission planning and replanning using Multi-Objective Evolutionary Algorithms (MOEAs).
3. Use of Multi-Criteria Decision Making (MCDM) methods based on operator preferences for the selection of the final plan.
4. Extension of QGroundControl, a GCS simulation framework, with an interface for mission planning, replanning and decision making.

References

[1]: Ramírez Atencia C. (2018). Automated Mission Planning and Decision Support Systems for Multiple Unmanned Aerial Vehicles, PhD Thesis, Airbus Defence & Space Savier project (Situational Awareness Virtual Environment)

Bibliography

Atencia, C. R., Del Ser, J., & Camacho, D. (2019). Weighted strategies to guide a multi-objective evolutionary algorithm for multi-UAV mission planning. Swarm and evolutionary computation, 44, 480-495.

Panizo-LLedot, A., Bello-Orgaz, G., & Camacho, D. (2019). A Multi-Objective Genetic Algorithm for detecting dynamic communities using a local search driven immigrant’s scheme. Future Generation Computer Systems.

Ramirez-Atencia, C., & Camacho, D. (2019). Constrained multi-objective optimization for multi-UAV planning. Journal of Ambient Intelligence and Humanized Computing, 10(6), 2467-2484.

Ramirez-Atencia, C., Bello-Orgaz, G., R-Moreno, M. D., & Camacho, D. (2017). Solving complex multi-UAV mission planning problems using multi-objective genetic algorithms. Soft Computing, 21(17), 4883-4900.

Waltz, E. L., & Buede, D. M. (1986). Data fusion and decision support for command and control. IEEE Transactions on Systems, Man, and Cybernetics, 16(6), 865-879.